InfoScale™ Cluster Server 9.0 Bundled Agents Reference Guide - AIX
- Introducing bundled agents
- Storage agents
- DiskGroup agent
- Notes for DiskGroup agent
- Sample configurations for DiskGroup agent
- DiskGroupSnap agent
- Notes for DiskGroupSnap agent
- Sample configurations for DiskGroupSnap agent
- Volume agent
- VolumeSet agent
- Sample configurations for VolumeSet agent
- LVMVG agent
- Notes for LVMVG agent
- Mount agent
- Sample configurations for Mount agent
- SFCache agent
- Network agents
- About the network agents
- IP agent
- NIC agent
- IPMultiNIC agent
- MultiNICA agent
- About the IPMultiNICB and MultiNICB agents
- IPMultiNICB agent
- Sample configurations for IPMultiNICB agent
- MultiNICB agent
- Sample configurations for MultiNICB agent
- DNS agent
- Agent notes for DNS agent
- About using the VCS DNS agent on UNIX with a secure Windows DNS server
- Sample configurations for DNS agent
- File share agents
- NFS agent
- NFSRestart agent
- Share agent
- About the Samba agents
- Notes for configuring the Samba agents
- SambaServer agent
- SambaShare agent
- NetBios agent
- Service and application agents
- Apache HTTP server agent
- Application agent
- Notes for Application agent
- Sample configurations for Application agent
- CoordPoint agent
- LPAR agent
- Notes for LPAR agent
- MemCPUAllocator agent
- MemCPUAllocator agent notes
- Process agent
- Usage notes for Process agent
- Sample configurations for Process agent
- ProcessOnOnly agent
- RestServer agent
- WPAR agent
- Infrastructure and support agents
- Testing agents
- Replication agents
About the updatepv action
The updatepv action updates the volume group's physical volumes (PV) information on all the other nodes in the cluster.
If a volume group's disk configuration is changed on a node, the updated information is not automatically propagated to other nodes of the cluster, and thus some or all of the ODM entries on those nodes become stale. This may cause the online operation of the volume group resource to fail.
You must run this action whenever there are changes in the disk configuration, such as addition, deletion, or replacement of physical volumes of the volume group. Running this action ensures that the other nodes are updated with the new information, which is used when the agent brings the volume group online. Use the following command to run this action.
# hares - action res_name updatepv - sys system_name
where system_name is the name of the node on which the disks were added to or removed from the volume group.
Running this action sends the updated physical volume information from the local node to all the other nodes. This information is stored in the file /var/VRTSvcs/log/tmp/resource_name.volume_group_name.pvid on all the nodes. If this file is present on that node, and if the SyncODM attribute is set to 1, then the online entry point uses the PVIDs from the file, exports the volume group, breaks the reservations on all these disks, and uses any one PVID to re-import the volume group. The file is deleted from that node after the volume group is successfully brought online on that node and the ODM is synchronized.
The SyncODM attribute must the set to 1 if the updatepv action has been executed for that volume group.
To ensure the high availability of the LVMVG resource, you must run the updatepv action immediately after adding, deleting, or replacing the physical volumes in the volume group. When the new disks are added or replaced in the volume group, these disks must be visible and have same PVID on all the cluster nodes.
You must run the updatepv action again for the nodes that were down when updatepv was last run.
Note:
The updatepv action does not support the GCO environment.